Climate and Development

Carolina Torreblanca

University of Pennsylvania

Global Development: Intermediate Topics in Politics, Policy, and Data

PSCI 3200 - Spring 2026

Agenda

  1. Climate and Development
  2. Dell, Jones & Olken (2012)
  3. Interpreting Logged Variables

Deadlines

  • Expanded Research Design due Wednesday, April 15 by 11:59pm via Slack
    • Make sure your html renders properly and you can see the plots if you open it on a web browser!

Climate and Development

The Planet Is Warming

Who Caused It

Should Climate Change Affect Development?

  • Agriculture: crop yields fall above optimal temperature thresholds
  • Labor productivity: heat stress reduces output, especially outdoor and manual work
  • Health: heat illness, expanded disease vectors (malaria, dengue)
  • Conflict: resource scarcity increases stress and violence
  • Infrastructure: extreme weather destroys capital

Who Suffers

Dell, Jones & Olken (2012)

Hot Countries Are Poor

What Do You Notice?

  • Hotter places tend to be poorer. Does that mean heat causes poverty?
  • What else differs between Nigeria and Norway besides temperature?
  • Colonization history, institutions, geography, disease burden, trade access…
  • The cross-country pattern cannot isolate the effect of temperature

Theory

If temperature does affect growth, where would we expect to see it most?

  • Places with less capacity to adapt: rain-fed agriculture, weak health systems, fragile institutions
  • So … poor countries

Data and Model

125 countries, 1950–2003. Temperature from gridded weather data. GDP growth from Penn World Tables.

\[\Delta \ln y_{it} = \beta_1 T_{it} + \beta_2 P_{it} + \alpha_i + \gamma_t + \epsilon_{it}\]

  • \(T_{it}\): mean temperature in country \(i\), year \(t\), relative to that country’s mean over the full sample (1950–2003)
  • \(\alpha_i\): country fixed effects
  • \(\gamma_t\): year fixed effects

The Identification Strategy

  • Country and Year FE absorb everything time-invariant for the country and year including “climate”
  • \(T_{it}\) measures was this year unusually hot for this country? Weather
  • Identification assumption: conditional on country and year FE, year-to-year temperature fluctuations are uncorrelated with other determinants of GDP growth (i.e., \(T_{it}\) is exogenous)

What the Assumption Allows and Forbids

Reading the Table

Col (1): no effect on average across all countries.

Col (2): temperature interacted with a poor country dummy. Large and significant (-1.655***).

Marginal effect in poor countries: β_temp + β_interaction = -1.4pp. Same logic as slopes().

What the Paper Argues

  • The rich-poor asymmetry suggests climate change will widen global inequality
  • Historical emissions came overwhelmingly from rich countries; damages fall on poor ones
  • These are effects on growth rates, not levels; compounding matters enormously
  • A 1.3pp drag sustained over decades means massive foregone development
  • Adaptation is possible but requires investment poor countries cannot self-finance

Interpreting Logged Variables

Why Log? Reason 1: Skewed Distributions

Why Log? Reason 2: A One-Unit Increase Is Not the Same

From $403 to $1,097

  • Log distance: 1 unit
  • Dollar jump: $694

From $22,026 to $59,874

  • Log distance: 1 unit
  • Dollar jump: $37,848



A $700 raise is a lot if you earn $400. Not so much if you earn $40,000. Log GDP measures gains proportionally.

Log Scale vs. Dollar Scale

  • Log: proportional changes matter (income, GDP, population)
  • Levels: absolute changes matter (temperature, vote share)
  • Modeling choice: Dell logs GDP, not temperature

Caution: Coefficients Mean Something Different

The familiar reading: \(\hat{\beta}\) = “a one-unit increase in \(X\) raises \(Y\) by \(\hat{\beta}\) units”

When \(Y\) is \(\ln Y\): \(\hat{\beta} \times 100 \approx\) % change in \(Y\)

When \(Y\) is \(\Delta \ln Y\) (already a growth rate): \(\hat{\beta}\) is directly a percentage point change

Interpretation also changes if \(X\) is logged